CLENQUEUEWRITEBUFFER(3clc)

📦
version 1.0~svn33624 · 3 versions
, in ubuntu jammy, ubuntu noble

💖 clEnqueueWriteBufferRect — the OpenCL manual page, bestie edition 💖

okay so you've stumbled into clEnqueueWriteBufferRect(3clc), which lives in the OpenCL Manual, and honestly? she's a whole vibe. this is the one that lets you yeet a 2D or 3D rectangular region out of your host memory and into a buffer object. no cap. let's get into it 🫶

✨ NAME ✨

clEnqueueWriteBufferRect — enqueue commands to write a 2D or 3D rectangular region to a buffer object from host memory. 🦋

✨ SYNOPSIS ✨

cl_int clEnqueueWriteBufferRect(cl_command_queue command_queue,
                                cl_mem buffer, cl_bool blocking_write,
                                const size_t * buffer_origin,
                                const size_t * host_origin,
                                const size_t *region,
                                size_t buffer_row_pitch,
                                size_t buffer_slice_pitch,
                                size_t host_row_pitch,
                                size_t host_slice_pitch, void *ptr,
                                cl_uint num_events_in_wait_list,
                                const cl_event *event_wait_list,
                                cl_event *event);

look at her. fourteen arguments. she's high maintenance and she knows it 💅

✨ PARAMETERS ✨

  • command_queue — the command-queue where this write command is gonna get queued up. important: command_queue and buffer have to be created with the same OpenCL context, or it's not gonna fly, bestie.
  • buffer — has to point at a valid buffer object. that's it, that's the rule. 🎀
  • blocking_write — the drama switch. does the write block or not:
    • if blocking_write is CL_TRUE, the OpenCL implementation copies the data that ptr points to and enqueues the write operation in the command-queue. the memory that ptr points to can be reused by the application once the clEnqueueWriteBufferRect call returns. so like, you're free 💖
    • if blocking_write is CL_FALSE, the OpenCL implementation will use ptr to do a nonblocking write. because the write is non-blocking, the implementation can just return immediately, no thoughts. but heads up: the memory ptr points to cannot be reused by the application after the call returns. the event argument gives you back an event object you can use to query the execution status of the write command. once the write command has completed, then the memory ptr points to can be reused. patience, queen. 🧚
  • buffer_origin — the (x, y, z) offset in the memory region associated with buffer. for a 2D rectangle region, the z value given by buffer_origin[2] should be 0. the offset in bytes is computed as:
buffer_origin[2] * buffer_slice_pitch + buffer_origin[1] * buffer_row_pitch + buffer_origin[0]
  • host_origin — the (x, y, z) offset in the memory region pointed to by ptr. for a 2D rectangle region, the z value given by host_origin[2] should be 0. the offset in bytes is computed as:
host_origin[2] * host_slice_pitch + host_origin[1] * host_row_pitch + host_origin[0]
  • region — the (width, height, depth) in bytes of the 2D or 3D rectangle being read or written. for a 2D rectangle copy, the depth value given by region[2] should be 1. ⭐
  • buffer_row_pitch — the length of each row in bytes to be used for the memory region associated with buffer. if buffer_row_pitch is 0, then buffer_row_pitch is computed as region[0]. she's self-sufficient like that.
  • buffer_slice_pitch — the length of each 2D slice in bytes to be used for the memory region associated with buffer. if buffer_slice_pitch is 0, then buffer_slice_pitch is computed as region[1] * buffer_row_pitch.
  • host_row_pitch — the length of each row in bytes to be used for the memory region pointed to by ptr. if host_row_pitch is 0, then host_row_pitch is computed as region[0].
  • host_slice_pitch — the length of each 2D slice in bytes to be used for the memory region pointed to by ptr. if host_slice_pitch is 0, then host_slice_pitch is computed as region[1] * host_row_pitch.
  • ptr — the pointer to buffer in host memory where data is to be written from. 📥
  • event_wait_list, num_events_in_wait_list — these two specify events that need to complete before this particular command can even be executed. the deets:
    • if event_wait_list is NULL, then this particular command does not wait on any event to complete.
    • if event_wait_list is NULL, then num_events_in_wait_list must be 0.
    • if event_wait_list is not NULL, the list of events pointed to by event_wait_list must be valid, and num_events_in_wait_list must be greater than 0.
    • the events specified in event_wait_list act as synchronization points.
    • the context associated with events in event_wait_list and command_queue must be the same.
    • the memory associated with event_wait_list can be reused or freed after the function returns. 🎉
  • event — returns an event object that identifies this particular write command, and can be used to query or queue a wait for this particular command to complete. event can be NULL, in which case it will not be possible for the application to query the status of this command or queue a wait for this command to complete. so like, your loss, but valid. 🙈

⚠️ NOTES ⚠️

okay this part is the serious one, don't skip it, bestie. 😭

calling clEnqueueWriteBufferRect to update the latest bits in a region of the buffer object, with the ptr argument value set to host_ptr and host_origin, buffer_origin values the same — where host_ptr is a pointer to the memory region specified when the buffer object being written is created with CL_MEM_USE_HOST_PTR — must meet the following requirements in order to avoid undefined behavior:

  • the host memory region given by (buffer_origin region) contains the latest bits when the enqueued write command begins execution.
  • the buffer object, or memory objects created from this buffer object, are not mapped.
  • the buffer object, or memory objects created from this buffer object, are not used by any command-queue until the write command has finished execution.
undefined behavior is not a cute look. read those three bullets again. 💀

💀 ERRORS 💀

clEnqueueWriteBufferRect returns CL_SUCCESS if the function is executed successfully. otherwise, she returns one of the following errors, and every single one is a different flavor of "ma'am":

  • CL_INVALID_COMMAND_QUEUE — if command_queue is not a valid command-queue.
  • CL_INVALID_CONTEXT — if the context associated with command_queue and buffer are not the same, or if the context associated with command_queue and events in event_wait_list are not the same.
  • CL_INVALID_MEM_OBJECT — if buffer is not a valid buffer object.
  • CL_INVALID_VALUE — if the region being read specified by (buffer_origin, region, buffer_row_pitch, buffer_slice_pitch) is out of bounds.
  • CL_INVALID_VALUE — if ptr is a NULL value.
  • CL_INVALID_VALUE — if any region array element is 0.
  • CL_INVALID_VALUE — if buffer_row_pitch is not 0 and is less than region[0].
  • CL_INVALID_VALUE — if host_row_pitch is not 0 and is less than region[0].
  • CL_INVALID_VALUE — if buffer_slice_pitch is not 0 and is less than region[1] * buffer_row_pitch and not a multiple of buffer_row_pitch.
  • CL_INVALID_VALUE — if host_slice_pitch is not 0 and is less than region[1] * host_row_pitch and not a multiple of host_row_pitch.
  • CL_INVALID_EVENT_WAIT_LIST — if event_wait_list is NULL and num_events_in_wait_list > 0, or event_wait_list is not NULL and num_events_in_wait_list is 0, or if event objects in event_wait_list are not valid events.
  • CL_MISALIGNED_SUB_BUFFER_OFFSET — if buffer is a sub-buffer object and the offset specified when the sub-buffer object is created is not aligned to CL_DEVICE_MEM_BASE_ADDR_ALIGN value for the device associated with queue.
  • CL_EXEC_STATUS_ERROR_FOR_EVENTS_IN_WAIT_LIST — if the read and write operations are blocking and the execution status of any of the events in event_wait_list is a negative integer value.
  • CL_MEM_OBJECT_ALLOCATION_FAILURE — if there is a failure to allocate memory for data store associated with buffer.
  • CL_INVALID_OPERATION — if clEnqueueWriteBufferRect is called on buffer which has been created with CL_MEM_HOST_READ_ONLY or CL_MEM_HOST_NO_ACCESS.
  • CL_OUT_OF_RESOURCES — if there is a failure to allocate resources required by the OpenCL implementation on the device.
  • CL_OUT_OF_HOST_MEMORY — if there is a failure to allocate resources required by the OpenCL implementation on the host.

📜 SPECIFICATION 📜

OpenCL Specification[1]

👀 SEE ALSO 👀

the whole friend group. iconic. 🔥

👑 AUTHORS 👑

The Khronos Group

Copyright © 2007-2011 The Khronos Group Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and/or associated documentation files (the "Materials"), to deal in the Materials without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Materials, and to permit persons to whom the Materials are furnished to do so, subject to the condition that this copyright notice and permission notice shall be included in all copies or substantial portions of the Materials.

📝 NOTES 📝

  1. OpenCL Specification page 75, section 5.2.2 - Reading, Writing and Copying Buffer Objects

that's the whole page, bestie, you ate it. 💅✨ this came to you from The Khronos Group, dated 01/14/2021, and it's CLENQUEUEWRITEBUFFER(3clc) — go forth and write some rectangles, and please, for the love of everything, keep command_queue and buffer in the same context. mwah 🫶🎀

📦 where this page comes from